What is Ground Improvement Testing?

The term ground improvement covers a range of techniques for improving the shear strength and settlement characteristics of the near surface soils, either through immediate compaction and densification, or through reinforcement with materials of high shear strength.

Whilst not providing an exhaustive list, this blog explores some of the different types of testing used to validate the range of ground improvement methods GeMech offer – vibro stone columns, dynamic compaction and rigid inclusion – their importance, and their applications.

We will also explore some of the common terms used when discussing the design and validation of ground improvement work for the purposes of ground bearing slab design.

Ground improvement testing can be differentiated between methods used for quality control checks of completed elements of the works, and those used for the estimation of long-term settlement performance.

Types of Testing Commonly Used on Ground Improvement Contracts

1. Plate Load Testing

Small diameter plate load tests – using test plate diameters ranging from 450mm up to 600mm – are used primarily for quality control checks, to validate the stiffness of vibro stone columns, or the load settlement response of a load transfer platform over rigid inclusions.

A short duration test method, the maximum applied load is typically 2.5 times working load, allowing an assessment to be made of the shallow soils’ immediate response to load application.

The depth of meaningful stress increase will be a function of test plate diameter, ranging from approximately 700mm to 900mm for 450mm and 600mm plates respectively (1.5B). Given this constraint on meaningful depth of stress increase it is clear that plate load tests can give no indication of the load/settlement performance of the improved soil mass, with the potential exception of very small shallow pad foundations. However, satisfactory test results can give confidence that the work has completed in accordance with the design requirements.

Whilst discussing plate load testing, it would be useful to consider two parameters often used in ground bearing slab design that are obtained from plate load tests, namely –

The procedure for carrying out an in-situ CBR test on site is described in BS1377-9:1990 section 4.3. It involves forcing an approximately 50mm diameter plate into the soil at a constant rate of 1 +/-0.2mm per minute, with time readings taken for each 0.25mm of penetration up to a maximum of 7.5mm. The small dimensional area of the CBR test plate is such that the stresses imposed during the test are limited to just the upper 100mm or so of soil. As a result, the recorded CBR value can give no meaningful indication of the settlement performance of the improved soil but can give useful information about the stiffness of the subgrade soils providing support to ground bearing slabs.

Modulus of Subgrade Reaction (ks)

TR34 Concrete Industrial Ground Floors, a guide to design and construction, provides the following guidance –

The measured modulus of sub-grade reaction is a function of the load applied to induce 1.25mm settlement of a 750mm diameter plate. A test plate of this diameter will induce meaningful stress increase over approximately the upper 1.15m of soil. In common with the CBR test, the recorded modulus of subgrade reaction value can give no indication of the settlement performance of the improved soil but can give useful information about the stiffness of the subgrade soils providing support to ground bearing slabs.

2. Zone Load Testing

Large scale zone load tests, or, for residential housing projects, dummy strip footing tests, are used to validate the load/settlement performance of the improved soil mass.

Owing to the complexities involved in providing sufficient reaction for larger scale tests, zone load tests carried out in the UK tend to use a plate size of up to 3.0m x 3.0m square, reducing to a1.5m long x 0.6m wide test plate for a dummy strip footing test.

For a typical vibro stone column design bearing pressure of 150kPa and allowing for advancing the zone load test to 1.5 x working load, a 3.0m x 3.0m square zone load test will require over 2,000kN of reaction, achieved either through the installation of reaction piles, or, more commonly, through the mobilisation of 200t + of kentledge. In cases such as this steel ballast weights would ordinarily be used as kentledge, although even then, taking unit weight of steel as 7.0t/m³, the height of the kentledge would be over 3.0m high, or a bespoke test frame would need to be constructed – no mean feat!

Given the complications and costs associated with the testing described above it is preferable to minimise the zone load test size as far as practical, which reduces the effectiveness of the testing method for validating the load/settlement performance of the improved soil, for the reasons illustrated below –

Figure 1. Idealised schematic of a 3.0m x 3.0m zone load test on a hypothetical site, which is underlain by an initial 4.5m of Made Ground over natural Clay soils.

Increases in effective stress are deemed ‘meaningful’ with respect to foundation settlements where they exceed 20% of the in-situ effective stress state, as estimated by the Boussinesq approximation at an approximate depth of 1.5 x foundation width – assuming that no significant plastic deformations occur.

Figure 2. Idealised schematic of a 4.0m x 4.0m x 1.0m deep working foundation on a hypothetical site, which is underlain by an initial 4.5m of Made Ground over natural Clay soils.

The contrast between the two conditions is clear. The 3.0m x 3.0m zone load test applied at surface level will measure the load settlement performance of the 4.5m depth of improved Made Ground. Whereas the meaningful increases in stress associated with the 4.0m x 4.0m working foundation placed at 1.0m below ground level will extend into the deeper natural Clays soils. Also, the performance of the working foundation will take some benefit from the positive effect of the overburden pressure above base of foundation level.

This consideration of the depth of meaningful effective stress increase is exacerbated when applied to wide loaded areas subject to uniformly distributed loads, where the depth of influence would be expected to be significantly deeper than that associated with a typical load test.

With all things considered it must be recognised that zone load testing is unlikely to provide a completely accurate indication of load/settlement performance of loaded slabs, or foundations which have a much larger footprint area than the test plate dimensions. Additionally, to obtain a true indication of the load/settlement performance of smaller foundations, it is advisable to progress the test from a similar depth to the proposed depth of the working foundations.

3. Geophysical Testing (CSW)

A novel method of testing and validating ground improvement work is the use of geophysical seismic methods, such as continuous surface wave testing.

For a more detailed description of this technique please visit https://www.soilsafe.co.uk/

A brief outline of the technique is described below –

Step 1 –

The equipment is brought to site in a 4×4 vehicle, and is set up as shown in the image above, with a vibration unit positioned at a known offset from an array of geophones positioned at surface level.

Step 2 –

The vibration unit is operated at a range of frequencies, and the response times of the surface waves are recorded at each geophone position. Using the geophone response time data, and the known distance between each geophone in the array, the velocity of the surface waves can be calculated – minimum strain shear stiffness can be calculated as a function of shear wave velocity. A typical profile of shear wave velocity recorded during CSW testing is given below.

The wavelength associated with each frequency can be attributed to soils at varying depth below survey level – both simple and advanced methods inversion methods are available for establishing a shear wave velocity profile with depth from the recorded data.

Step 3 –

A profile of minimum strain shear stiffness modulus is derived from shear wave velocity graph.

Step 4 –

A strain softening function is used to estimate an appropriate value of operational shear stiffness, based on estimates of shear strain occurring in the stressed layers of soil below base of foundation level. Typically, appropriate levels of operational stiffness are in the order of 10% – 30% of the minimum strain shear stiffness values estimated from shear wave velocity.

It is true to say that the inversion problem has no single unique solution, and that no specific strain softening function has been devised appropriate to all soil types, or, perhaps more importantly, to inhomogeneous layered soils. However, for the reasons provided above where zone load testing is discussed in greater detail, the results obtained are potentially no more or less accurate than the load/settlement relationship suggested by physical testing techniques.

The major benefit of using CSW testing in lieu of zone load testing is cost benefit. For the cost of a single zone load test, it is possible to arrange three, perhaps four days’ of CSW testing, yielding between thirty and sixty unique small strain shear stiffness profiles, allowing a lot more of the improved soil to be analysed. A reasonable compromise could be the use of CSW testing in combination with a single Zone Load test, allowing the results of the two tests methods to be compared.

For best results from CSW testing site ‘noise’ must be minimised. Best outcomes are achieved when no plant is used on site whilst the testing is in progress, ensuring that the recorded geophone data is not detrimentally affected by competing surface waves from other sources.

5. Integrity Testing

Specific to ground improvement using displacement or replacement augering rigid inclusion methods, integrity testing can be used to assesses the structural integrity and continuity of the unreinforced inclusions, identifying defects such as cracks, voids, and unwanted inclusions.

The ASIRI (2012) document – the leading design guidance for rigid inclusion design and implementation – recommends that displacement rigid inclusions are integrity tested wherever they are installed at spacings closer than four times rigid inclusion diameter.

In UK practice it is common to install rigid inclusions from top of load transfer platform level, and to reduce the rigid inclusions whilst the concrete is wet. The load transfer platform is then recompacted over the reduced inclusion head, leaving them hidden upon completion of the rigid inclusion construction phase, hindering the successful completion of integrity testing after the inclusions have cured.

6. Concrete Cube Testing

Again, specific to ground improvement using displacement or replacement augering rigid inclusion methods, concrete cube testing can be used to measure the strength of the concrete to firm the rigid inclusions, and to confirm it meets the design requirements.

It should be noted that when buying a designated mix from a QRMC contractor the design strength is effectively guaranteed by the supplier, who must take regular samples and keep a record of the achieved strengths for the designated mixes they supply. With this being the case, concrete cube testing becomes more an exercise in identity testing – ensuring the correct grade of concrete has been used – rather than strength validation

It should also be recognised that the entire process of concrete sampling, cube making, storage and curing needs to be strictly controlled. In our experience, low strength results are more often the result of poor procedure than defective concrete. In our experience, very few sites have either the facilities or necessary expertise to carry out the procedure correctly. As a result, GeMech use external accredited materials testing subcontractors to sample, make and test concrete cubes on our behalf.

Want to know more? Contact us today to find out how we can help you with your ground improvement project.